U.S. patent application number 16/971737 was filed with the patent office on 2021-02-04 for method, system, and computer program product for applying deep learning analysis to financial device usage.
The applicant listed for this patent is Visa International Service Association. Invention is credited to Nikan Chavoshi, Mahashweta Das, Hao Yang.
Application Number | 20210035141 16/971737 |
Document ID | / |
Family ID | 1000005206441 |
Filed Date | 2021-02-04 |
![](/patent/app/20210035141/US20210035141A1-20210204-D00000.png)
![](/patent/app/20210035141/US20210035141A1-20210204-D00001.png)
![](/patent/app/20210035141/US20210035141A1-20210204-D00002.png)
![](/patent/app/20210035141/US20210035141A1-20210204-D00003.png)
![](/patent/app/20210035141/US20210035141A1-20210204-D00004.png)
![](/patent/app/20210035141/US20210035141A1-20210204-D00005.png)
![](/patent/app/20210035141/US20210035141A1-20210204-D00006.png)
![](/patent/app/20210035141/US20210035141A1-20210204-D00007.png)
![](/patent/app/20210035141/US20210035141A1-20210204-M00001.png)
![](/patent/app/20210035141/US20210035141A1-20210204-P00001.png)
United States Patent
Application |
20210035141 |
Kind Code |
A1 |
Das; Mahashweta ; et
al. |
February 4, 2021 |
Method, System, and Computer Program Product for Applying Deep
Learning Analysis to Financial Device Usage
Abstract
Described are a system, method, and computer program product for
applying deep learning analysis to predict and automatically
respond to predicted changes in financial device primacy for a
financial device holder. The method includes receiving transaction
data representative of a plurality of transactions between the
financial device holder and a merchant. The method also includes
generating time series data based on the transaction data and
generating a predictive model configured to: (i) receive an input
of time-interval-based transaction data; and (ii) output a
probability of primary financial device primacy change, the
predictive model trained based on historic transaction data. The
method further includes determining a probability of primary
financial device primacy change for the financial device holder by
applying the predictive model to the time series data. The method
further includes, generating at least one communication to at least
one issuer and/or the financial device holder.
Inventors: |
Das; Mahashweta; (Sunnyvale,
CA) ; Chavoshi; Nikan; (Santa Clara, CA) ;
Yang; Hao; (San Jose, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Visa International Service Association |
San Francisco |
CA |
US |
|
|
Family ID: |
1000005206441 |
Appl. No.: |
16/971737 |
Filed: |
February 23, 2018 |
PCT Filed: |
February 23, 2018 |
PCT NO: |
PCT/US2018/019314 |
371 Date: |
August 21, 2020 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G06Q 30/016 20130101;
G06N 3/0472 20130101; G06Q 40/12 20131203; G06N 3/08 20130101; G06N
3/0445 20130101; G06Q 40/02 20130101; G06Q 30/0215 20130101; G06N
3/0454 20130101 |
International
Class: |
G06Q 30/02 20060101
G06Q030/02; G06N 3/08 20060101 G06N003/08; G06N 3/04 20060101
G06N003/04; G06Q 40/02 20060101 G06Q040/02; G06Q 40/00 20060101
G06Q040/00; G06Q 30/00 20060101 G06Q030/00 |
Claims
1. A computer-implemented method for applying deep learning
analysis to predict and automatically respond to predicted changes
in financial device primacy for a financial device holder, the
method comprising: receiving, with at least one processor,
transaction data representative of a plurality of transactions
between the financial device holder and at least one merchant over
a first time interval; generating, with at least one processor,
time series data based on the transaction data, the time series
data comprising: (i) a plurality of subintervals over the first
time interval; and (ii) a set of generated statistical parameters
for each of the plurality of subintervals, the set of statistical
parameters based at least partially on at least one of the
following: transaction count; transaction amount; transaction date;
transaction day of week; transaction time of day; transaction
merchant type; or any combination thereof; generating, with at
least one processor, a predictive model configured to: (i) receive
an input of time-interval-based transaction data; and (ii) output a
probability of primary financial device primacy change, the
predictive model trained based on historic transaction data;
determining, with at least one processor, a probability of primary
financial device primacy change for the financial device holder by
applying the predictive model to the time series data; and in
response to determining that the probability of primary financial
device primacy change satisfies a threshold, generating, with at
least one processor and based at least partially on the probability
of primary financial device primacy change for the financial device
holder, at least one communication to at least one issuer and/or
the financial device holder.
2. The method of claim 1, further comprising: determining, with at
least one processor, one or more applicable issuer promotions for a
current primary financial device of the financial device holder by
retrieving issuer promotion data from an issuer promotion database;
and transmitting, with at least one processor, the at least one
communication to the financial device holder, the at least one
communication comprising the one or more applicable issuer
promotions for the current primary financial device to discourage
the financial device holder from changing their primary financial
device.
3. The method of claim 1, further comprising: receiving, with at
least one processor, new transaction data representative of a
plurality of transactions between the financial device holder and
the at least one merchant over a second time interval; generating,
with at least one processor, new time series data based on the new
transaction data, the new time series data comprising a plurality
of subintervals over the second time interval; determining, with at
least one processor, a new probability of primary financial device
primacy change for the financial device holder by applying the
predictive model to the new time series data and based at least
partially on at least one prior probability prediction determined
from application of the predictive model to prior-generated time
series data; and transmitting, with at least one processor and
based at least partially on the new probability of primary
financial device primacy change for the financial device holder, at
least one of the following: (i) the at least one communication to
the at least one issuer indicating a likelihood of the financial
device holder changing their primary financial device; (ii) the at
least one communication to the financial device holder to
discourage the financial device holder from changing their primary
financial device or encourage the financial device holder to change
their primary financial device; or (iii) any combination
thereof.
4. The method of claim 3, further comprising updating the
predictive model based at least partially on the new transaction
data.
5. The method of claim 1, wherein financial device data of a
current primary financial device of the financial device holder for
the first time interval is stored in a database in association with
issuer correspondence data.
6. The method of claim 5, further comprising transmitting, with at
least one processor and based at least partially on the issuer
correspondence data, the at least one communication to the at least
one issuer, the at least one issuer comprising an issuer of the
current primary financial device, and the at least one
communication indicating that the financial device holder is likely
to have a new primary financial device in a subsequent time
interval.
7. The method of claim 5, further comprising transmitting, with at
least one processor and based at least partially on the issuer
correspondence data, the at least one communication to the at least
one issuer, the at least one issuer comprising an issuer of a
non-primary financial device, and the at least one communication
indicating that the financial device holder may have a new primary
financial device in a subsequent time interval, in response to
determining that the probability of primary financial device
primacy change for the financial device holder is more likely than
not.
8. The method of claim 1, wherein financial device data of a
current primary financial device of the financial device holder for
the first time interval is stored in a database in association with
financial device holder correspondence data.
9. The method of claim 8, further comprising, in response to
detecting a transaction request for the primary financial device,
transmitting, with at least one processor and based at least
partially on the financial device holder correspondence data, the
at least one communication to the financial device holder, the at
least one communication comprising an offer or reward.
10. The method of claim 1, wherein primary financial device primacy
is defined by the ratio of the number of transactions for a given
financial device over the total number of transactions for a
financial device holder being greater than or equal to a predefined
value greater than 0.6.
11. A system for applying deep learning analysis to predict and
automatically respond to predicted changes in financial device
primacy for a financial device holder, the system comprising at
least one server computer including at least one processor, the at
least one server computer programmed and/or configured to: receive
transaction data representative of a plurality of transactions
between the financial device holder and at least one merchant over
a first time interval; generate time series data based on the
transaction data, the time series data comprising: (i) a plurality
of subintervals over the first time interval; and (ii) a set of
generated statistical parameters for each of the plurality of
subintervals, the set of statistical parameters based at least
partially on at least one of the following: transaction count;
transaction amount; transaction date; transaction day of week;
transaction time of day; transaction merchant type; or any
combination thereof; generate a predictive model configured to: (i)
receive an input of time-interval-based transaction data; and (ii)
output a probability of primary financial device primacy change,
the predictive model trained based on historic transaction data;
determine a probability of primary financial device primacy change
for the financial device holder by applying the predictive model to
the time series data; and in response to determining that the
probability of primary financial device primacy change satisfies a
threshold, generate, based at least partially on the probability of
primary financial device primacy change for the financial device
holder, at least one communication to at least one issuer and/or
the financial device holder.
12. The system of claim 11, the at least one server computer
further programmed and/or configured to: determine one or more
applicable issuer promotions for a current primary financial device
of the financial device holder by retrieving issuer promotion data
from an issuer promotion database; and transmit the at least one
communication to the financial device holder, the at least one
communication comprising the one or more applicable issuer
promotions for the current primary financial device to discourage
the financial device holder from changing their primary financial
device.
13. The system of claim 11, the at least one server computer
further programmed and/or configured to: receive new transaction
data representative of a plurality of transactions between the
financial device holder and the at least one merchant over a second
time interval; generate new time series data based on the new
transaction data, the new time series data comprising a plurality
of subintervals over the second time interval; determine a new
probability of primary financial device primacy change for the
financial device holder by applying the predictive model to the new
time series data and based at least partially on at least one prior
probability prediction determined from application of the
predictive model to prior-generated time series data; and transmit,
based at least partially on the new probability of primary
financial device primacy change for the financial device holder, at
least one of the following: (i) the at least one communication to
the at least one issuer indicating a likelihood of the financial
device holder changing their primary financial device; (ii) the at
least one communication to the financial device holder to
discourage the financial device holder from changing their primary
financial device or encourage the financial device holder to change
their primary financial device; or (iii) any combination
thereof.
14. The system of claim 13, the at least one server computer
further programmed and/or configured to update the predictive model
based at least partially on the new transaction data.
15. The system of claim 11, wherein financial device data of a
current primary financial device of the financial device holder for
the first time interval is stored in a database in association with
issuer correspondence data.
16. The system of claim 15, the at least one server computer
further programmed and/or configured to transmit, based at least
partially on the issuer correspondence data, the at least one
communication to the at least one issuer, the at least one issuer
comprising an issuer of the current primary financial device, and
the at least one communication indicating that the financial device
holder is likely to have a new primary financial device in a
subsequent time interval.
17. The system of claim 15, the at least one server computer
further programmed and/or configured to transmit, based at least
partially on the issuer correspondence data, the at least one
communication to the at least one issuer, the at least one issuer
comprising an issuer of a non-primary financial device, and the at
least one communication indicating that the financial device holder
may have a new primary financial device in a subsequent time
interval, in response to determining that the probability of
primary financial device primacy change for the financial device
holder is more likely than not.
18. The system of claim 11, wherein financial device data of a
current primary financial device of the financial device holder for
the first time interval is stored in a database in association with
financial device holder correspondence data.
19. The method of claim 18, the at least one server computer
further programmed and/or configured to, in response to detecting a
transaction request for the primary financial device, transmit,
based at least partially on the financial device holder
correspondence data, the at least one communication to the
financial device holder, the at least one communication comprising
an offer or reward.
20. The method of claim 11, wherein primary financial device
primacy is defined by the ratio of the number of transactions for a
given financial device over the total number of transactions for a
financial device holder being greater than or equal to a predefined
value greater than 0.6.
21.-30. (canceled)
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application is the United States national phase of
International Application No. PCT/US2018/019314 filed Feb. 23,
2018, the disclosure of which is hereby incorporated by reference
in its entirety.
BACKGROUND
1. Technical Field
[0002] Disclosed embodiments relate generally to a system, method,
and computer program product for applying deep learning analysis to
predict and automatically respond to predicted changes in financial
device primacy for a financial device holder, and In some
non-limiting embodiments or aspects, to a system, method, and
computer program product for generating and applying a predictive
model to generated time series data of transactions to predict
financial device primacy changes and react accordingly.
2. Technical Considerations
[0003] Consumers use financial devices (e.g., credit cards, debit
cards, etc.) to complete financial transactions with merchants. The
financial device that a consumer is most likely to use to complete
a given financial transaction may be referred to as a primary
financial device, or with regard to the use of payment cards, a
top-of-wallet (TOW) card. The market for financial devices, e.g.,
the payment card market, is incredibly competitive, and issuers
actively compete for their respective financial device to have
primacy. There are many possible factors for why a particular
financial device may become a primary financial device, including
financial device incentives (e.g., cashback/rewards programs), the
reliability and convenience of fraud detection, and the
availability of other financial devices. It is desirable for
issuers to understand consumer habits for financial device usage in
order to design targeted campaigns to increase market share in the
financial device market. If an issuer can anticipate when a
consumer will change their primary financial device, the issuer can
offer rewards or perks to ensure the change occurs (if the current
primary financial device is not associated with the issuer) or
prevent the change from occurring (if the current primary financial
device is associated with the issuer). However, it is difficult for
an individual issuer to meaningfully interpret and predict consumer
behavior when the information available to an issuer is largely
limited to their own market share.
[0004] There is a need in the art to reliably predict consumer
financial device usage and determine when a consumer is likely to
change their primary financial device. There is a further need to
make such predictions automatically and substantially in real-time,
to allow for short response times to take action, and to trigger
relevant, targeted communications.
SUMMARY
[0005] Accordingly, and generally, provided is an improved system,
computer-implemented method, and computer program product for
applying deep learning analysis to predict and automatically
respond to predicted changes in financial device primacy for a
financial device holder. Preferably, provided is an improved
system, computer-implemented method, and computer program product
for receiving transaction data representative of a plurality of
transactions between the financial device holder and at least one
merchant, and generating time series data based on the transaction
data. Preferably, provided is an improved system,
computer-implemented method, and computer program product for
generating a predictive model configured to: (i) receive an input
of time-interval-based transaction data; and (ii) output a
probability of primary financial device primacy change. Preferably,
provided is an improved system, computer-implemented method, and
computer program product for determining a probability of primary
financial device primacy change for the financial device holder by
applying the predictive model to the time series data, and
generating at least one communication to at least one issuer and/or
the financial device holder.
[0006] According to some non-limiting embodiments or aspects,
provided is a computer-implemented method for applying deep
learning analysis to predict and automatically respond to predicted
changes in financial device primacy for a financial device holder.
The method includes receiving, with at least one processor,
transaction data representative of a plurality of transactions
between the financial device holder and at least one merchant over
a first time interval. The method also includes generating, with at
least one processor, time series data based on the transaction
data. The time series data includes: (i) a plurality of
subintervals over the first time interval; and (ii) a set of
generated statistical parameters for each of the plurality of
subintervals. The set of statistical parameters is based at least
partially on at least one of the following: transaction count;
transaction amount; transaction date; transaction day of week;
transaction time of day; transaction merchant type; or any
combination thereof. The method further includes generating, with
at least one processor, a predictive model configured to: (i)
receive an input of time-interval-based transaction data; and (ii)
output a probability of primary financial device primacy change,
the predictive model trained based on historic transaction data.
The method further includes determining, with at least one
processor, a probability of primary financial device primacy change
for the financial device holder by applying the predictive model to
the time series data. The method further includes, in response to
determining that the probability of primary financial device
primacy change satisfies a threshold, generating, with at least one
processor and based at least partially on the probability of
primary financial device primacy change for the financial device
holder, at least one communication to at least one issuer and/or
the financial device holder.
[0007] In some some non-limiting embodiments or aspects, the method
may include determining, with at least one processor, one or more
applicable issuer promotions for a current primary financial device
of the financial device holder by retrieving issuer promotion data
from an issuer promotion database. The method may also include
transmitting, with at least one processor, the at least one
communication to the financial device holder. The at least one
communication may include the one or more applicable issuer
promotions for the current primary financial device to discourage
the financial device holder from changing their primary financial
device.
[0008] In some non-limiting embodiments or aspects, the method may
include receiving, with at least one processor, new transaction
data representative of a plurality of transactions between the
financial device holder and the at least one merchant over a second
time interval. The method may also include generating, with at
least one processor, new time series data based on the new
transaction data. The new time series data may include a plurality
of subintervals over the second time interval. The method may
further include determining, with at least one processor, a new
probability of primary financial device primacy change for the
financial device holder by applying the predictive model to the new
time series data and based at least partially on at least one prior
probability prediction determined from application of the
predictive model to prior-generated time series data. The method
may further include transmitting, with at least one processor and
based at least partially on the new probability of primary
financial device primacy change for the financial device holder, at
least one of the following: (i) the at least one communication to
the at least one issuer indicating a likelihood of the financial
device holder changing their primary financial device; (ii) the at
least one communication to the financial device holder to
discourage the financial device holder from changing their primary
financial device or encourage the financial device holder to change
their primary financial device; or (iii) any combination
thereof.
[0009] In some non-limiting embodiments or aspects, the method may
include updating the predictive model based at least partially on
the new transaction data. Financial device data of a current
primary financial device of the financial device holder for the
first time interval may be stored in a database in association with
issuer correspondence data. The method may also include
transmitting, with at least one processor and based at least
partially on the issuer correspondence data, the at least one
communication to the at least one issuer. The at least one issuer
may include an issuer of the current primary financial device, and
the at least one communication may indicate that the financial
device holder is likely to have a new primary financial device in a
subsequent time interval. The at least one issuer may also include
an issuer of a non-primary financial device, and the at least one
communication may indicate that the financial device holder may
have a new primary financial device in a subsequent time interval,
in response to determining that the probability of primary
financial device primacy change for the financial device holder is
more likely than not.
[0010] In some non-limiting embodiments or aspects, financial
device data of a current primary financial device of the financial
device holder for the first time interval may be stored in a
database in association with financial device holder correspondence
data. The method may include, in response to detecting a
transaction request for the primary financial device, transmitting,
with at least one processor and based at least partially on the
financial device holder correspondence data, the at least one
communication to the financial device holder. The at least one
communication may include an offer or reward. Primary financial
device primacy may be defined by the ratio of the number of
transactions for a given financial device over the total number of
transactions for a financial device holder being greater than or
equal to a predefined value greater than 0.6.
[0011] According to some non-limiting embodiments or aspects,
provided is a system for applying deep learning analysis to predict
and automatically respond to predicted changes in financial device
primacy for a financial device holder. The system includes at least
one server computer including at least one processor. The at least
one server computer is programmed and/or configured to receive
transaction data representative of a plurality of transactions
between the financial device holder and at least one merchant over
a first time interval. The server computer is also programmed
and/or configured to generate time series data based on the
transaction data. The time series data includes: (i) a plurality of
subintervals over the first time interval; and (ii) a set of
generated statistical parameters for each of the plurality of
subintervals. The set of statistical parameters is based at least
partially on at least one of the following: transaction count;
transaction amount; transaction date; transaction day of week;
transaction time of day; transaction merchant type; or any
combination thereof. The server computer is further programmed
and/or configured to generate a predictive model configured to: (i)
receive an input of time-interval-based transaction data; and (ii)
output a probability of primary financial device primacy change.
The predictive model is trained based on historic transaction data.
The server computer is further programmed and/or configured to
determine a probability of primary financial device primacy change
for the financial device holder by applying the predictive model to
the time series data. The server computer is further programmed
and/or configured to, in response to determining that the
probability of primary financial device primacy change satisfies a
threshold, generate, based at least partially on the probability of
primary financial device primacy change for the financial device
holder, at least one communication to at least one issuer and/or
the financial device holder.
[0012] In some non-limiting embodiments or aspects, the at least
one server computer may be programmed and/or configured to
determine one or more applicable issuer promotions for a current
primary financial device of the financial device holder by
retrieving issuer promotion data from an issuer promotion database.
The at least one server computer may also be programmed and/or
configured to transmit the at least one communication to the
financial device holder, the at least one communication including
the one or more applicable issuer promotions for the current
primary financial device to discourage the financial device holder
from changing their primary financial device.
[0013] In some non-limiting embodiments or aspects, the at least
one server computer may be programmed and/or configured to receive
new transaction data representative of a plurality of transactions
between the financial device holder and the at least one merchant
over a second time interval. The at least one server computer may
also be programmed and/or configured to generate new time series
data based on the new transaction data. The new time series data
may include a plurality of subintervals over the second time
interval. The at least one server computer may further be
programmed and/or configured to determine a new probability of
primary financial device primacy change for the financial device
holder by applying the predictive model to the new time series data
and based at least partially on at least one prior probability
prediction determined from application of the predictive model to
prior-generated time series data. The at least one server computer
may further be programmed and/or configured to transmit, based at
least partially on the new probability of primary financial device
primacy change for the financial device holder, at least one of the
following: (i) the at least one communication to the at least one
issuer indicating a likelihood of the financial device holder
changing their primary financial device; (ii) the at least one
communication to the financial device holder to discourage the
financial device holder from changing their primary financial
device or encourage the financial device holder to change their
primary financial device; or (iii) any combination thereof.
[0014] In some non-limiting embodiments or aspects, the at least
one server computer may be programmed and/or configured to update
the predictive model based at least partially on the new
transaction data. Financial device data of a current primary
financial device of the financial device holder for the first time
interval may be stored in a database in association with issuer
correspondence data. The at least one server computer may also be
programmed and/or configured to transmit, based at least partially
on the issuer correspondence data, the at least one communication
to the at least one issuer. The at least one issuer may include an
issuer of the current primary financial device, and the at least
one communication indicating that the financial device holder is
likely to have a new primary financial device in a subsequent time
interval. The at least one issuer may also include an issuer of a
non-primary financial device, and the at least one communication
may indicate that the financial device holder may have a new
primary financial device in a subsequent time interval, in response
to determining that the probability of primary financial device
primacy change for the financial device holder is more likely than
not.
[0015] In some non-limiting embodiments or aspects, financial
device data of a current primary financial device of the financial
device holder for the first time interval may be stored in a
database in association with financial device holder correspondence
data. The at least one server computer may be programmed and/or
configured to, in response to detecting a transaction request for
the primary financial device, transmit, based at least partially on
the financial device holder correspondence data, the at least one
communication to the financial device holder, the at least one
communication including an offer or reward. Primary financial
device primacy may be defined by the ratio of the number of
transactions for a given financial device over the total number of
transactions for a financial device holder being greater than or
equal to a predefined value greater than 0.6.
[0016] According some non-limiting embodiments or aspects, provided
is a computer program product for applying deep learning analysis
to predict and automatically respond to predicted changes in
financial device primacy for a financial device holder. The
computer program product includes at least one non-transitory
computer-readable medium including program instructions that, when
executed by at least one processor, cause the at least one
processor to receive transaction data representative of a plurality
of transactions between the financial device holder and at least
one merchant over a first time interval. The program instructions
are also programmed and/or configured to cause the at least one
processor to generate time series data based on the transaction
data. The time series data includes: (i) a plurality of
subintervals over the first time interval; and (ii) a set of
generated statistical parameters for each of the plurality of
subintervals. The set of statistical parameters are based at least
partially on at least one of the following: transaction count;
transaction amount; transaction date; transaction day of week;
transaction time of day; transaction merchant type; or any
combination thereof. The program instructions are further
programmed and/or configured to cause the at least one processor to
generate a predictive model configured to: (i) receive an input of
time-interval-based transaction data; and (ii) output a probability
of primary financial device primacy change, the predictive model
trained based on historic transaction data. The program
instructions are further programmed and/or configured to cause the
at least one processor to determine a probability of primary
financial device primacy change for the financial device holder by
applying the predictive model to the time series data. The program
instructions are programmed and/or configured to cause the at least
one processor to, in response to determining that the probability
of primary financial device primacy change satisfies a threshold,
generate, based at least partially on the probability of primary
financial device primacy change for the financial device holder, at
least one communication to at least one issuer and/or the financial
device holder.
[0017] In some non-limiting embodiments or aspects, the program
instructions may be programmed and/or configured to cause the at
least one processor to determine one or more applicable issuer
promotions for a current primary financial device of the financial
device holder by retrieving issuer promotion data from an issuer
promotion database. The program instructions may also be programmed
and/or configured to cause the at least one processor to transmit
the at least one communication to the financial device holder. The
at least one communication may include the one or more applicable
issuer promotions for the current primary financial device to
discourage the financial device holder from changing their primary
financial device.
[0018] In some non-limiting embodiments or aspects, the program
instructions may be programmed and/or configured to cause the at
least one processor to receive new transaction data representative
of a plurality of transactions between the financial device holder
and the at least one merchant over a second time interval. The
program instructions may also be programmed and/or configured to
cause the at least one processor to generate new time series data
based on the new transaction data. The new time series data may
include a plurality of subintervals over the second time interval.
The program instructions may further be programmed and/or
configured to cause the at least one processor to determine a new
probability of primary financial device primacy change for the
financial device holder by applying the predictive model to the new
time series data and based at least partially on at least one prior
probability prediction determined from application of the
predictive model to prior-generated time series data. The program
instructions may also be programmed and/or configured to cause the
at least one processor to transmit, based at least partially on the
new probability of primary financial device primacy change for the
financial device holder, at least one of the following: (i) the at
least one communication to the at least one issuer indicating a
likelihood of the financial device holder changing their primary
financial device; (ii) the at least one communication to the
financial device holder to discourage the financial device holder
from changing their primary financial device or encourage the
financial device holder to change their primary financial device;
or (iii) any combination thereof.
[0019] In some non-limiting embodiments or aspects, the program
instructions may be programmed and/or configured to cause the at
least one processor to update the predictive model based at least
partially on the new transaction data. Financial device data of a
current primary financial device of the financial device holder for
the first time interval may be stored in a database in association
with issuer correspondence data. The program instructions may also
be programmed and/or configured to cause the at least one processor
to transmit, based at least partially on the issuer correspondence
data, the at least one communication to the at least one issuer.
The at least one issuer may include an issuer of the current
primary financial device, and the at least one communication may
indicate that the financial device holder is likely to have a new
primary financial device in a subsequent time interval. The at
least one issuer may include an issuer of a non-primary financial
device, and the at least one communication may indicate that the
financial device holder may have a new primary financial device in
a subsequent time interval, in response to determining that the
probability of primary financial device primacy change for the
financial device holder is more likely than not.
[0020] In some non-limiting embodiments or aspects, financial
device data of a current primary financial device of the financial
device holder for the first time interval may be stored in a
database in association with financial device holder correspondence
data. The program instructions may be programmed and/or configured
to cause the at least one processor to, in response to detecting a
transaction request for the primary financial device, transmit,
based at least partially on the financial device holder
correspondence data, the at least one communication to the
financial device holder, the at least one communication including
an offer or reward. Primary financial device primacy may be defined
by the ratio of the number of transactions for a given financial
device over the total number of transactions for a financial device
holder being greater than or equal to a predefined value greater
than 0.6.
[0021] Further non-limiting embodiments or aspects of the present
invention are set forth in the following numbered clauses:
[0022] Clause 1: A computer-implemented method for applying deep
learning analysis to predict and automatically respond to predicted
changes in financial device primacy for a financial device holder,
the method comprising: receiving, with at least one processor,
transaction data representative of a plurality of transactions
between the financial device holder and at least one merchant over
a first time interval; generating, with at least one processor,
time series data based on the transaction data, the time series
data comprising: (i) a plurality of subintervals over the first
time interval; and (ii) a set of generated statistical parameters
for each of the plurality of subintervals, the set of statistical
parameters based at least partially on at least one of the
following: transaction count; transaction amount; transaction date;
transaction day of week; transaction time of day; transaction
merchant type; or any combination thereof; generating, with at
least one processor, a predictive model configured to: (i) receive
an input of time-interval-based transaction data; and (ii) output a
probability of primary financial device primacy change, the
predictive model trained based on historic transaction data;
determining, with at least one processor, a probability of primary
financial device primacy change for the financial device holder by
applying the predictive model to the time series data; and in
response to determining that the probability of primary financial
device primacy change satisfies a threshold, generating, with at
least one processor and based at least partially on the probability
of primary financial device primacy change for the financial device
holder, at least one communication to at least one issuer and/or
the financial device holder.
[0023] Clause 2: The method of clause 1, further comprising:
determining, with at least one processor, one or more applicable
issuer promotions for a current primary financial device of the
financial device holder by retrieving issuer promotion data from an
issuer promotion database; and transmitting, with at least one
processor, the at least one communication to the financial device
holder, the at least one communication comprising the one or more
applicable issuer promotions for the current primary financial
device to discourage the financial device holder from changing
their primary financial device.
[0024] Clause 3: The method of clause 1 or 2, further comprising:
receiving, with at least one processor, new transaction data
representative of a plurality of transactions between the financial
device holder and the at least one merchant over a second time
interval; generating, with at least one processor, new time series
data based on the new transaction data, the new time series data
comprising a plurality of subintervals over the second time
interval; determining, with at least one processor, a new
probability of primary financial device primacy change for the
financial device holder by applying the predictive model to the new
time series data and based at least partially on at least one prior
probability prediction determined from application of the
predictive model to prior-generated time series data; and
transmitting, with at least one processor and based at least
partially on the new probability of primary financial device
primacy change for the financial device holder, at least one of the
following: (i) the at least one communication to the at least one
issuer indicating a likelihood of the financial device holder
changing their primary financial device; (ii) the at least one
communication to the financial device holder to discourage the
financial device holder from changing their primary financial
device or encourage the financial device holder to change their
primary financial device; or (iii) any combination thereof.
[0025] Clause 4: The method of any of clauses 1-3, further
comprising updating the predictive model based at least partially
on the new transaction data.
[0026] Clause 5: The method of any of clauses 1-4, wherein
financial device data of a current primary financial device of the
financial device holder for the first time interval is stored in a
database in association with issuer correspondence data.
[0027] Clause 6: The method of any of clauses 1-5, further
comprising transmitting, with at least one processor and based at
least partially on the issuer correspondence data, the at least one
communication to the at least one issuer, the at least one issuer
comprising an issuer of the current primary financial device, and
the at least one communication indicating that the financial device
holder is likely to have a new primary financial device in a
subsequent time interval.
[0028] Clause 7: The method of any of clauses 1-6, further
comprising transmitting, with at least one processor and based at
least partially on the issuer correspondence data, the at least one
communication to the at least one issuer, the at least one issuer
comprising an issuer of a non-primary financial device, and the at
least one communication indicating that the financial device holder
may have a new primary financial device in a subsequent time
interval, in response to determining that the probability of
primary financial device primacy change for the financial device
holder is more likely than not.
[0029] Clause 8: The method of any of clauses 1-7, wherein
financial device data of a current primary financial device of the
financial device holder for the first time interval is stored in a
database in association with financial device holder correspondence
data.
[0030] Clause 9: The method of any of clauses 1-8, further
comprising, in response to detecting a transaction request for the
primary financial device, transmitting, with at least one processor
and based at least partially on the financial device holder
correspondence data, the at least one communication to the
financial device holder, the at least one communication comprising
an offer or reward.
[0031] Clause 10: The method of any of clauses 1-9, wherein primary
financial device primacy is defined by the ratio of the number of
transactions for a given financial device over the total number of
transactions for a financial device holder being greater than or
equal to a predefined value greater than 0.6.
[0032] Clause 11: A system for applying deep learning analysis to
predict and automatically respond to predicted changes in financial
device primacy for a financial device holder, the system comprising
at least one server computer including at least one processor, the
at least one server computer programmed and/or configured to:
receive transaction data representative of a plurality of
transactions between the financial device holder and at least one
merchant over a first time interval; generate time series data
based on the transaction data, the time series data comprising: (i)
a plurality of subintervals over the first time interval; and (ii)
a set of generated statistical parameters for each of the plurality
of subintervals, the set of statistical parameters based at least
partially on at least one of the following: transaction count,
transaction amount, transaction date, transaction day of week,
transaction time of day, transaction merchant type, or any
combination thereof; generate a predictive model configured to: (i)
receive an input of time-interval-based transaction data; and (ii)
output a probability of primary financial device primacy change,
the predictive model trained based on historic transaction data;
determine a probability of primary financial device primacy change
for the financial device holder by applying the predictive model to
the time series data; and in response to determining that the
probability of primary financial device primacy change satisfies a
threshold, generate, based at least partially on the probability of
primary financial device primacy change for the financial device
holder, at least one communication to at least one issuer and/or
the financial device holder.
[0033] Clause 12: The system of clause 11, the at least one server
computer further programmed and/or configured to: determine one or
more applicable issuer promotions for a current primary financial
device of the financial device holder by retrieving issuer
promotion data from an issuer promotion database; and transmit the
at least one communication to the financial device holder, the at
least one communication comprising the one or more applicable
issuer promotions for the current primary financial device to
discourage the financial device holder from changing their primary
financial device.
[0034] Clause 13: The system of clause 11 or 12, the at least one
server computer further programmed and/or configured to: receive
new transaction data representative of a plurality of transactions
between the financial device holder and the at least one merchant
over a second time interval; generate new time series data based on
the new transaction data, the new time series data comprising a
plurality of subintervals over the second time interval; determine
a new probability of primary financial device primacy change for
the financial device holder by applying the predictive model to the
new time series data and based at least partially on at least one
prior probability prediction determined from application of the
predictive model to prior-generated time series data; and transmit,
based at least partially on the new probability of primary
financial device primacy change for the financial device holder, at
least one of the following: (i) the at least one communication to
the at least one issuer indicating a likelihood of the financial
device holder changing their primary financial device; (ii) the at
least one communication to the financial device holder to
discourage the financial device holder from changing their primary
financial device or encourage the financial device holder to change
their primary financial device, or (iii) any combination
thereof.
[0035] Clause 14: The system of any of clauses 11-13, the at least
one server computer further programmed and/or configured to update
the predictive model based at least partially on the new
transaction data.
[0036] Clause 15: The system of any of clauses 11-14, wherein
financial device data of a current primary financial device of the
financial device holder for the first time interval is stored in a
database in association with issuer correspondence data.
[0037] Clause 16: The system of any of clauses 11-15, the at least
one server computer further programmed and/or configured to
transmit, based at least partially on the issuer correspondence
data, the at least one communication to the at least one issuer,
the at least one issuer comprising an issuer of the current primary
financial device, and the at least one communication indicating
that the financial device holder is likely to have a new primary
financial device in a subsequent time interval.
[0038] Clause 17: The system of any of clauses 11-16, the at least
one server computer further programmed and/or configured to
transmit, based at least partially on the issuer correspondence
data, the at least one communication to the at least one issuer,
the at least one issuer comprising an issuer of a non-primary
financial device, and the at least one communication indicating
that the financial device holder may have a new primary financial
device in a subsequent time interval, in response to determining
that the probability of primary financial device primacy change for
the financial device holder is more likely than not.
[0039] Clause 18: The system of any of clauses 11-17, wherein
financial device data of a current primary financial device of the
financial device holder for the first time interval is stored in a
database in association with financial device holder correspondence
data.
[0040] Clause 19: The method of any of clauses 11-18, the at least
one server computer further programmed and/or configured to, in
response to detecting a transaction request for the primary
financial device, transmit, based at least partially on the
financial device holder correspondence data, the at least one
communication to the financial device holder, the at least one
communication comprising an offer or reward.
[0041] Clause 20: The method of any of clauses 11-19, wherein
primary financial device primacy is defined by the ratio of the
number of transactions for a given financial device over the total
number of transactions for a financial device holder being greater
than or equal to a predefined value greater than 0.6.
[0042] Clause 21: A computer program product for applying deep
learning analysis to predict and automatically respond to predicted
changes in financial device primacy for a financial device holder,
the computer program product comprising at least one non-transitory
computer-readable medium including program instructions that, when
executed by at least one processor, cause the at least one
processor to: receive transaction data representative of a
plurality of transactions between the financial device holder and
at least one merchant over a first time interval; generate time
series data based on the transaction data, the time series data
comprising: (i) a plurality of subintervals over the first time
interval; and (ii) a set of generated statistical parameters for
each of the plurality of subintervals, the set of statistical
parameters based at least partially on at least one of the
following: transaction count; transaction amount; transaction date;
transaction day of week; transaction time of day; transaction
merchant type; or any combination thereof; generate a predictive
model configured to: (i) receive an input of time-interval-based
transaction data; and (ii) output a probability of primary
financial device primacy change, the predictive model trained based
on historic transaction data; determine a probability of primary
financial device primacy change for the financial device holder by
applying the predictive model to the time series data; and in
response to determining that the probability of primary financial
device primacy change satisfies a threshold, generate, based at
least partially on the probability of primary financial device
primacy change for the financial device holder, at least one
communication to at least one issuer and/or the financial device
holder.
[0043] Clause 22: The computer program product of clause 21, the
program instructions being further programmed and/or configured to
cause the at least one processor to: determine one or more
applicable issuer promotions for a current primary financial device
of the financial device holder by retrieving issuer promotion data
from an issuer promotion database; and transmit the at least one
communication to the financial device holder, the at least one
communication comprising the one or more applicable issuer
promotions for the current primary financial device to discourage
the financial device holder from changing their primary financial
device.
[0044] Clause 23: The computer program product of clause 21 or 22,
the program instructions being further programmed and/or configured
to cause the at least one processor to: receive new transaction
data representative of a plurality of transactions between the
financial device holder and the at least one merchant over a second
time interval; generate new time series data based on the new
transaction data, the new time series data comprising a plurality
of subintervals over the second time interval; determine a new
probability of primary financial device primacy change for the
financial device holder by applying the predictive model to the new
time series data and based at least partially on at least one prior
probability prediction determined from application of the
predictive model to prior-generated time series data; and transmit,
based at least partially on the new probability of primary
financial device primacy change for the financial device holder, at
least one of the following: (i) the at least one communication to
the at least one issuer indicating a likelihood of the financial
device holder changing their primary financial device; (ii) the at
least one communication to the financial device holder to
discourage the financial device holder from changing their primary
financial device or encourage the financial device holder to change
their primary financial device; or (iii) any combination
thereof.
[0045] Clause 24: The computer program product of any of clauses
21-23, the program instructions being further programmed and/or
configured to cause the at least one processor to update the
predictive model based at least partially on the new transaction
data.
[0046] Clause 25: The computer program product of any of clauses
21-24, wherein financial device data of a current primary financial
device of the financial device holder for the first time interval
is stored in a database in association with issuer correspondence
data.
[0047] Clause 26: The computer program product of any of clauses
21-25, the program instructions being further programmed and/or
configured to cause the at least one processor to transmit, based
at least partially on the issuer correspondence data, the at least
one communication to the at least one issuer, the at least one
issuer comprising an issuer of the current primary financial
device, and the at least one communication indicating that the
financial device holder is likely to have a new primary financial
device in a subsequent time interval.
[0048] Clause 27: The computer program product of any of clauses
21-26, the program instructions being further programmed and/or
configured to cause the at least one processor to transmit, based
at least partially on the issuer correspondence data, the at least
one communication to the at least one issuer, the at least one
issuer comprising an issuer of a non-primary financial device, and
the at least one communication indicating that the financial device
holder may have a new primary financial device in a subsequent time
interval, in response to determining that the probability of
primary financial device primacy change for the financial device
holder is more likely than not.
[0049] Clause 28: The computer program product of any of clauses
21-27, wherein financial device data of a current primary financial
device of the financial device holder for the first time interval
is stored in a database in association with financial device holder
correspondence data.
[0050] Clause 29: The computer program product of any of clauses
21-28, the program instructions being further programmed and/or
configured to cause the at least one processor to, in response to
detecting a transaction request for the primary financial device,
transmit, based at least partially on the financial device holder
correspondence data, the at least one communication to the
financial device holder, the at least one communication comprising
an offer or reward.
[0051] Clause 30: The computer program product of any of clauses
21-29, wherein primary financial device primacy is defined by the
ratio of the number of transactions for a given financial device
over the total number of transactions for a financial device holder
being greater than or equal to a predefined value greater than
0.6.
[0052] These and other features and characteristics of the present
invention, as well as the methods of operation and functions of the
related elements of structures and the combination of parts and
economies of manufacture, will become more apparent upon
consideration of the following description and the appended claims
with reference to the accompanying drawings, all of which form a
part of this specification, wherein like reference numerals
designate corresponding parts in the various figures. It is to be
expressly understood, however, that the drawings are for the
purpose of illustration and description only and are not intended
as a definition of the limits of the invention. As used in the
specification and the claims, the singular form of "a," "an," and
"the" include plural referents unless the context clearly dictates
otherwise.
BRIEF DESCRIPTION OF THE DRAWINGS
[0053] Additional advantages and details of the invention are
explained in greater detail below with reference to the exemplary
embodiments that are illustrated in the accompanying figures, in
which:
[0054] FIG. 1 is a schematic diagram of one non-limiting embodiment
or aspect of a system and method for applying deep learning
analysis to predict and automatically respond to predicted changes
in financial device primacy for a financial device holder;
[0055] FIG. 2 is a flow diagram of one non-limiting embodiment or
aspect of a system and method for applying deep learning analysis
to predict and automatically respond to predicted changes in
financial device primacy for a financial device holder;
[0056] FIG. 3 is a flow diagram of one non-limiting embodiment or
aspect of a system and method for applying deep learning analysis
to predict and automatically respond to predicted changes in
financial device primacy for a financial device holder;
[0057] FIG. 4 is a flow diagram of one non-limiting embodiment or
aspect of a system and method for applying deep learning analysis
to predict and automatically respond to predicted changes in
financial device primacy for a financial device holder;
[0058] FIG. 5 is a state transition diagram of one non-limiting
embodiment or aspect of a system and method for applying deep
learning analysis to predict and automatically respond to predicted
changes in financial device primacy for a financial device
holder;
[0059] FIG. 6 is a schematic diagram of one non-limiting embodiment
or aspect of a system and method for applying deep learning
analysis to predict and automatically respond to predicted changes
in financial device primacy for a financial device holder; and
[0060] FIG. 7 is a schematic diagram of one non-limiting embodiment
or aspect of a system and method for applying deep learning
analysis to predict and automatically respond to predicted changes
in financial device primacy for a financial device holder.
DETAILED DESCRIPTION
[0061] For purposes of the description hereinafter, the terms
"upper," "lower," "right," "left," "vertical," "horizontal," "top,"
"bottom," "lateral," "longitudinal," and derivatives thereof shall
relate to the disclosure as it is oriented in the drawing figures.
However, it is to be understood that the disclosure may assume
various alternative variations and step sequences, except where
expressly specified to the contrary. It is also to be understood
that the specific devices and processes illustrated in the attached
drawings, and described in the following specification, are simply
exemplary embodiments of the disclosure. Hence, specific dimensions
and other physical characteristics related to the embodiments
disclosed herein are not to be considered as limiting. Also, it
should be understood that any numerical range recited herein is
intended to include all sub-ranges subsumed therein. For example, a
range of "1 to 10" is intended to include all sub-ranges between
(and including) the recited minimum value of 1 and the recited
maximum value of 10, that is, having a minimum value equal to or
greater than 1 and a maximum value of equal to or less than 10.
[0062] As used herein, the terms "communication" and "communicate"
refer to the receipt or transfer of one or more signals, messages,
commands, or other type of data. For one unit (e.g., any device,
system, or component thereof) to be in communication with another
unit means that the one unit is able to directly or indirectly
receive data from and/or transmit data to the other unit. This may
refer to a direct or indirect connection that is wired and/or
wireless in nature. Additionally, two units may be in communication
with each other even though the data transmitted may be modified,
processed, relayed, and/or routed between the first and second
unit. For example, a first unit may be in communication with a
second unit even though the first unit passively receives data and
does not actively transmit data to the second unit. As another
example, a first unit may be in communication with a second unit if
an intermediary unit processes data from one unit and transmits
processed data to the second unit. It will be appreciated that
numerous other arrangements are possible.
[0063] As used herein, the term "transaction service provider" may
refer to an entity that receives transaction authorization requests
from merchants or other entities and provides guarantees of
payment, in some cases through an agreement between the transaction
service provider and an issuer institution. The terms "transaction
service provider" and "transaction service provider system" may
also refer to one or more computer systems operated by or on behalf
of a transaction service provider, such as a transaction processing
server executing one or more software applications. A transaction
processing server may include one or more processors and, in some
non-limiting embodiments or aspects, may be operated by or on
behalf of a transaction service provider.
[0064] As used herein, the term "account identifier" may include
one or more Personal Account Numbers (PANs), tokens, or other
identifiers associated with a customer account. The term "token"
may refer to an identifier that is used as a substitute or
replacement identifier for an original account identifier, such as
a PAN. Account identifiers may be alphanumeric or any combination
of characters and/or symbols. Tokens may be associated with a PAN
or other original account identifier in one or more databases such
that they can be used to conduct a transaction without directly
using the original account identifier. In some examples, an
original account identifier, such as a PAN, may be associated with
a plurality of tokens for different individuals or purposes. An
issuer institution may be associated with a bank identification
number (BIN) or other unique identifier that uniquely identifies it
among other issuer institutions.
[0065] As used herein, the term "merchant" may refer to an
individual or entity that provides goods and/or services, or access
to goods and/or services, to customers based on a transaction, such
as a payment transaction. The term "merchant" or "merchant system"
may also refer to one or more computer systems operated by or on
behalf of a merchant, such as a server computer executing one or
more software applications. A "point-of-sale (POS) system," as used
herein, may refer to one or more computers and/or peripheral
devices used by a merchant to engage in payment transactions with
customers, including one or more card readers, near-field
communication (NFC) receivers, RFID receivers, and/or other
contactless transceivers or receivers, contact-based receivers,
payment terminals, computers, servers, input devices, and/or other
like devices that can be used to initiate a payment
transaction.
[0066] As used herein, the term "mobile device" may refer to one or
more portable electronic devices configured to communicate with one
or more networks. As an example, a mobile device may include a
cellular phone (e.g., a smartphone or standard cellular phone), a
portable computer (e.g., a tablet computer, a laptop computer,
etc.), a wearable device (e.g., a watch, pair of glasses, lens,
clothing, and/or the like), a personal digital assistant (PDA),
and/or other like devices. The term "client device," as used
herein, refers to any electronic device that is configured to
communicate with one or more servers or remote devices and/or
systems. A client device may include a mobile device, a
network-enabled appliance (e.g., a network-enabled television,
refrigerator, thermostat, and/or the like), a computer, a POS
system, and/or any other device or system capable of communicating
with a network.
[0067] As used herein, the term "financial device" may refer to a
portable payment card (e.g., a credit or debit card), a gift card,
a smartcard, smart media, a payroll card, a healthcare card, a
wrist band, a machine-readable medium containing account
information, a keychain device or fob, an RFID transponder, a
retailer discount or loyalty card, a mobile device executing an
electronic wallet application, a PDA, a security card, an access
card, a wireless terminal, and/or a transponder, as examples. The
financial device may include a volatile or a non-volatile memory to
store information, such as an account identifier or a name of the
account holder. The financial device may store account credentials
locally on the device, in digital or non-digital representation, or
may facilitate accessing account credentials stored in a medium
that is accessible by the financial device in a connected
network.
[0068] As used herein, the term "primacy" may refer to favoring a
particular financial device as a primary financial device for
completing financial transactions. If a user has more than one
financial device, a financial device that has primacy may be
considered to be used by the user more often than the other
alternative financial devices. If a user only has one financial
device, that financial device may be said to have primacy. If all
financial devices of a user are used relatively equally, it may be
said that no financial device of the user has primacy. As used
herein, the term "primacy change" or "change in primacy" may refer
to one or more of the following conditions: a user having a
financial device with primacy in a first time no longer has a
financial device with primacy in a second time; a user having no
financial device with primacy in a first time has a financial
device with primacy in a second time; or, a user having a financial
device with primacy in a first time has a different financial
device with primacy in a second time. A lack of primacy change, or
a lack of change in primacy, may refer to one or more of the
following conditions: a user having a financial device with primacy
in a first time has the same financial device with primacy in a
second time; or, a user having no financial device with primacy in
a first time continues to have no financial device with primacy in
a second time. A change in primacy may be due to removing a
financial device from use or adding a financial device to use for a
user. A financial device that has primacy as the default financial
device in an electronic wallet (or e-wallet) or physical wallet may
be referred to as "top-of-wallet" (TOW).
[0069] As used herein, the term "server" may refer to or include
one or more processors or computers, storage devices, or similar
computer arrangements that are operated by or facilitate
communication and processing for multiple parties in a network
environment, such as the Internet, although it will be appreciated
that communication may be facilitated over one or more public or
private network environments and that various other arrangements
are possible. Further, multiple computers, e.g., servers, or other
computerized devices, e.g., POS devices, directly or indirectly
communicating in the network environment may constitute a "system,"
such as a merchant's POS system. Reference to "a server" or "a
processor," as used herein, may refer to a previously-recited
server and/or processor that is recited as performing a previous
step or function, a different server and/or processor, and/or a
combination of servers and/or processors. For example, as used in
the specification and the claims, a first server and/or a first
processor that is recited as performing a first step or function
may refer to the same or different server and/or a processor
recited as performing a second step or function.
[0070] The term "account data," as used herein, refers to any data
concerning one or more accounts for one or more users. Account data
may include, for example, one or more account identifiers, user
identifiers, transaction histories, balances, credit limits, issuer
institution identifiers, and/or the like.
[0071] In non-limiting embodiments or aspects of the present
disclosure, provided is a unique system architecture to analyze
consumer behavior in real-time as transactions are occurring and
being processed by a transaction service provider. By basing such
an analysis of consumer behavior on ongoing transaction data, and
by leveraging the processing position of the transaction service
provider, the described non-limiting embodiments provide the
benefit of reducing time delays and costs associated with
traditional methods of surveys or qualitative analysis. Individual
issuers that lack the infrastructure or market share to derive
meaningful data from post-processed transaction data can leverage
the much greater market share of a transaction service provider to
analyze a much wider sample of consumers. Furthermore, by using
deep learning techniques and neural networks as a basis for a
predictive model, non-limiting embodiments of the system may
anticipate changes in primary financial devices before they occur,
which provides significant cost and time savings over pure
reactionary systems. Additionally, probabilities that are output
from the predictive model may be compared to actual changes in
primacy as they occur, as a means of ongoing error correction,
which provides the benefit of recursive and iterative improvement
to the underlying model. This allows the computer system to
self-improve as more transaction data is collected for consumers
over time. Moreover, non-limiting embodiments directed to
communicating automatic notifications and promotions/offers to user
devices provide an improvement to e-wallet technology.
[0072] With specific reference to FIG. 1, and in some non-limiting
embodiments or aspects of the disclosure, provided is a system 100
for applying deep learning analysis to predict and automatically
respond to predicted changes in financial device primacy for a
financial device holder. In particular, a financial device holder
102 may have one or more financial devices 104 that may be used to
complete financial transactions with one or more merchants 106. A
given financial device 104 may be used at a merchant POS terminal
108, which is configured to transmit a transaction authorization
request to a transaction processing server 110. The transaction
processing server 110 may be associated with a transaction service
provider 111. The authorization request includes transaction data,
which may include, but is not limited to, transaction amount,
transaction description, transaction time, transaction date,
financial device data, financial device holder data, transaction
location, merchant identifier, merchant type, financial device
issuer, acquirer for merchant account, and/or the like. The
transaction data may be stored in a transaction data database 112
that is communicatively connected to the transaction processing
server 110. The transaction processing server 110 may act as a
transaction handler positioned between an issuer processor and an
acquirer processor to complete the transaction as a credit from the
financial device holder's account to the merchant's account. It
will be appreciated that many configurations are possible.
[0073] With further reference to FIG. 1, and in some non-limiting
embodiments or aspects of the disclosure, the system also may
include a predictive modeling server 114 configured to generate a
predictive model for predicting changes in financial device primacy
for one or more financial device holders 102. The predictive
modeling server 114 may be the same server as the transaction
processing server 110. The predictive modeling server 114 may be
configured to generate time series data based on the transaction
data received over a first time interval (e.g., one week, one
month, and/or the like). The transaction data is collected from a
plurality of transactions between a financial device holder 102 and
one or more merchants 106. The time series data may be divided into
a plurality of subintervals (e.g., days, weeks, and/or the like).
For each subinterval, the time series data may include a set of
statistical parameters for each subinterval that are based on the
transaction data, including, but not limited to, number of
transactions per subinterval, number of transactions on weekdays
per subinterval, number of transactions on weekends per
subinterval, number of transactions during night per subinterval,
number of transactions during morning per subinterval, number of
transactions during afternoon per subinterval, number of
transactions during evening per subinterval, total amount of
transactions per subinterval, average (e.g., statistical mean,
median, mode) transaction amount per subinterval, count or amount
of transactions with merchants in a designated merchant category,
and/or the like. The set of statistical parameters for each
subinterval may be concatenated to form a data map of vectors of
parameters representing the entire sample time interval. It will be
appreciated that many configurations are possible.
[0074] With further reference to FIG. 1, and in some non-limiting
embodiments or aspects of the disclosure, the predictive modeling
server 114 may output a probability of financial device primacy
change for one or more financial device holders 102. This
probability may be compared to a predetermined threshold, such as
p>0.50, p.gtoreq.0.60, p.gtoreq.0.70, and/or the like. Based on
this comparison, if the probability of financial device primacy
change for a financial device holder meets and/or exceeds the
predetermined threshold, a communication server 118 may generate
and transmit a communication to the financial device holder 102, to
an issuer of the primary financial device 122, or to an issuer of a
non-primary financial device 124. A financial device holder 102 may
receive the communication via a user communication device 120,
e.g., a mobile device. The communication server 118 may be the same
server as the predictive modeling server 114 and/or the transaction
processing server 110. A communication to a financial device holder
102 may include an offer/reward to encourage the user to change
their primary financial device, or to discourage the user from
changing their primary financial device. A communication to an
issuer of the primary financial device 122 may include an alert
that the financial device holder 102 is likely to change their
primary financial device, to allow the primary financial device
issuer 122 to encourage the financial device holder 102 to maintain
their primary financial device. A communication to an issuer of a
non-primary financial device 124 may include an alert that the
financial device holder 102 is likely to change their primary
financial device, to allow the non-primary financial device issuer
124 to encourage the financial device holder 102 to switch their
primary financial device. It will be appreciated that many
configurations are possible.
[0075] With further reference to FIG. 1, and in some non-limiting
embodiments or aspects of the disclosure, the system may further
include an issuer interface 126, e.g., a web portal, an application
programming interface (API), an app, and/or the like, to allow
issuers 122, 124 to communicate with an issuer management server
128, to edit and store issuer promotion data in a promotion
database 130. Issuer interfaces 126 may also be the means of
receiving communications from the communication server 118. The
issuer management server 128 may be the same server as the
communication server 118, predictive modeling server 114, and/or
the transaction processing server 110. These servers 110, 114, 118,
128 and their associated databases 112, 116, 130 may be associated
with and/or managed by the transaction service provider 111. Before
issuing a communication to a financial device holder 102 based on a
likelihood of financial device primacy change, the issuer
management server 128 may determine one or more applicable issuer
promotions for the financial device holder 102 by retrieving issuer
promotion data from the promotion database 130. Based on preset
offer campaigns, rewards campaigns, rebate campaigns, and/or the
like, the communication to the financial device holder 102 may
include one or more applicable issuer promotions to encourage or
discourage the financial device holder 102 from changing their
primary financial device. For example, when a transaction request
is being processed for a primary financial device, the financial
device holder 102 of which being likely to change their primary
financial device, the issuer management server 128 may identify a
promotional offer applicable to the purchase and transmit a
communication to a communication device 120 of the financial device
holder 102 to reinforce the purchase behavior. It will be
appreciated that the depicted system may continually receive new
transaction data, generate new time series data, determine new
probabilities of primary financial device primacy change, and
generate and transmit new communications. It will be appreciated
that many configurations are possible.
[0076] With specific reference to FIG. 2, and in some non-limiting
embodiments or aspects of the disclosure, provided is a method 200
for applying deep learning analysis to predict and automatically
respond to predicted changes in financial device primacy for a
financial device holder. The method 200 may be carried out by one
or more servers in the non-limiting exemplary system of FIG. 1. For
ease of reference in describing the method 200, they may be
individually or collectively referred to as "the processor" herein
below. At step 202, the processor receives transaction data
representative of a plurality of transactions between a given
financial device holder and at least one merchant over a first time
interval. At step 204, the processor generates time series data
based on the transaction data. The time series data may include a
plurality of subintervals over the first time interval and a set of
generated statistical parameters for each of the plurality of
subintervals. At step 206, based on the transaction data, the
processor may determine a financial device with primacy. This
primary financial device may be automatically determined as a
financial device that is used in more transactions than other
financial devices. It may also be determined as a financial device
that is used for a greater volume of transaction value than other
financial devices. It may also be determined as a financial device
that is used for a number of transactions that meet or exceed a
target threshold, such as 50%, 60%, or 70% of total transactions.
It will be appreciated that many configurations are possible.
[0077] With further reference to FIG. 2, and in some non-limiting
embodiments or aspects of the disclosure, at step 208, the primary
financial device data may be stored in association with
correspondence data (e.g., email address, phone number, user ID,
etc.) of the financial device holder and/or an issuer, such as the
issuer of the primary financial device, or an issuer of another
financial device of the financial device holder. At step 210, the
processor may generate a predictive model configured to receive an
input of time-interval-based transaction data and to output a
probability of primary financial device primacy change. The
predictive model may be trained based on historic transaction data.
At step 212, the processor may determine a probability of primary
financial device primacy change for the financial device holder by
applying the predictive model to the time series data generated
from the collected transaction data. At step 214, this probability
may be compared to a predetermined threshold. If the probability
satisfies the threshold (e.g., meets and/or passes the threshold),
the processor may generate and transmit a communication to the
financial device holder and/or an issuer, at step 216. After steps
212, 214, and/or 216, the processor may return to step 202 to
receive new transaction data from a subsequent (e.g., second,
third, fourth, etc.) time interval. Subsequently, step 204 may be
repeated to generate new time series data, step 212 may be repeated
to determine a new probability of primary financial device primacy
change, and step 216 may be repeated to send new communications. It
will be appreciated that many configurations are possible.
[0078] With specific reference to FIGS. 2-3, and in particular
reference to FIG. 3, in some non-limiting embodiments or aspects of
the disclosure, provided is a method 200 for applying deep learning
analysis to predict and automatically respond to predicted changes
in financial device primacy for a financial device holder. Depicted
are additional steps for execution by the processor after step 214,
particularly including step 218, in which the processor determines
one or more applicable issuer promotions from an issuer promotional
campaign. This may be retrieved from a promotion database as
configured prior by issuers. The offer may be applicable to the
current primary financial device. At step 220, the processor may
transmit at least one communication to the financial device holder,
the communication including an offer and/or reward. For example, an
offer may be a coupon for reduced cost of a future transactions,
which may automatically be applied upon processing the future
transaction. By way of further example, the reward may be a
monetary reward, a coupon code transmitted to the financial device
holder, and/or the like. The promotion may be related to the
current primary financial device, to discourage the financial
device holder from changing their primary financial device. The
promotion may be related to another financial device, to encourage
the financial device holder to change their primary financial
device. It will be appreciated that many configurations are
possible.
[0079] With specific reference to FIGS. 2-4, and in particular
reference to FIG. 4, in some non-limiting embodiments or aspects of
the disclosure, provided is a method 200 for applying deep learning
analysis to predict and automatically respond to predicted changes
in financial device primacy for a financial device holder. Depicted
are additional steps for execution by the processor after step 214,
particularly including step 222, in which the processor determines
if a stored issuer corresponds to the issuer of the current primary
financial device. At step 224, the processor may generate and
transmit a communication to an issuer notifying of a probable
change in financial device primacy. For the issuer of the primary
financial device, this communication would afford the issuer time
to encourage the financial device holder not to change their
primary financial device. For issuers of non-primary financial
devices, this communication would afford the issuers time to
further encourage the financial device holder to switch away from
their current primary financial device. It will be appreciated that
many configurations are possible.
[0080] With specific reference to FIG. 5, and in some non-limiting
embodiments or aspects of the disclosure, a change in primacy of a
financial device from one time interval to the next may be
represented using the principles of a finite state machine. For
example, each cardholder U may complete n transactions (n.gtoreq.0)
in a sample time interval T, by using one or more of k number of
financial devices C (represented as the set {C.sub.1, C.sub.2, . .
. C.sub.k}). The same cardholder U may complete n' transactions
(n'.gtoreq.0) in a subsequent sample time interval T', using one or
more of their financial devices ({C.sub.1, C.sub.2, . . .
C.sub.k}). As may be defined in some non-limiting embodiments or
aspects, there are four states in any given time interval: (1)
there exists a primary financial device C.sub.i in the first
interval T; (2) there exists a primary financial device C.sub.j in
the second time interval T'; (3) there is no primary financial
device C in a given time interval; (4) there are no transactions Y
in a given time interval. These four states are represented by
nodes in the depicted finite state machine diagram. In some
non-limiting embodiments or aspects, there are three transitions
between these states that may be deemed to be a change in financial
device primacy: (1) the financial device holder transitions from
primary financial device C.sub.i to primary financial device
C.sub.j, or vice versa; (2) the financial device holder transitions
from primary financial device C.sub.i to having no primary
financial device C, or vice versa; (3) the financial device holder
transitions from primary financial device C.sub.i to primary
financial device C.sub.j through a period of no transactions Y, or
vice versa. Transitions between states from a first time interval
(current state) to a second time interval (next state) are depicted
in the finite state machine diagram as arrows between nodes. The
transitions are labeled as "{input}/output". The "input" is
represented by "{prior state, next state}", and the "output" is
represented by either a "1" (representing a change in financial
device primacy) or "0" (representing no change in financial device
primacy). For example, the transition connecting current state
C.sub.j to next state C.sub.i labeled "{*, C.sub.i}/1" indicates
that if the previous state was null (e.g.,"*") and the next state
is C.sub.j, then there is a primacy change (e.g., "1"). By way of
further example, the transition connecting current state Y to next
state C.sub.j labeled "{C.sub.i, C.sub.j}/1" indicates that if the
previous state was C.sub.i and the next state is C.sub.j, then
there is a change in primacy (e.g., "1"). This logic may be used to
label the time series data with primacy changes from one time
interval or subinterval to another.
[0081] With specific reference to FIG. 6, and in some non-limiting
embodiments or aspects of the disclosure, depicted is a schematic
diagram of a neural network architecture 300 for a predictive model
for predicting changes in financial device primacy. A recurrent
neural network (RNN) may be used when analyzing sequential
dependencies in data. Long short-term memory (LSTM), a class of RNN
with sophisticated recurrent hidden and gated units, particularly
may be applied to identify hidden long-term sequential
dependencies. A convolutional neural network (CNN) may be used to
extract features from raw time series data for activity/action
recognition. As depicted in FIG. 6, a CNN may be employed as the
based predictive model for analyzing time-series formatted
transaction data. A CNN is made up of a number of algorithmic
layers, including an input layer 302, a convolution layer 304, a
pooling layer 306, a fully connected rectified linear unit (RELU)
layer 308, and an output layer 310. The convolution layer 304
generalizes the input to extract repeated patterns. The pooling
layer 306 determines the prominent information from the convolution
output. The convolution layer 304 and pooling layer 306 may be
repeated a number of times. The fully connected layer 308
identifies features within the processed data. Stacking the fully
connected layers 308 helps the model make better decisions since
each layer is utilized to identify hidden features.
[0082] With further reference to FIG. 6, and in some non-limiting
embodiments or aspects, the following formula may represent an
interval from a financial device holder's transaction
activities:
I.sub.i .di-elect cons. .sup.hw (Formula 1)
where w is the width of interval I.sub.i determined from the number
of statistical parameters (e.g., features) being analyzed, and
where height h depends on the number of subintervals within the
interval. For example, for a financial device holder whose
transactions are being analyzed at a weekly resolution (e.g.,
subinterval is one week), if each interval contains 14 features per
week and the financial device holder has transacted for a month
(assuming 1 month=4 weeks), the dimensions of the interval for the
financial device holder would be 4.times.14. In this model, padding
is not needed to make the subintervals in the data map uniform,
since all subintervals have the same size.
[0083] In view of this, each subinterval's feature vector may be
represented as w.sub.i. With this notation, an interval with m
weeks may be represented as:
I.sub.i .di-elect cons. w.sub.1 .sym. w.sub.2 .sym. . . . .sym.
w.sub.m (Formula 2)
The symbol ".sym." denotes concatenation of each week's feature
vectors, e.g., parameters of weekly transaction activities, to
represent the financial device holder's transaction history over
the entire sample time period.
[0084] The input 302 to the CNN predictive model for determining a
probability of financial device primacy change for a set of n users
is a set of d intervals collected from the n users. The various
layers of the CNN model may be applied for each of the d intervals
of n users to produce a series of p.sub.n probabilities. The d
intervals form the input layer 302. From the input layer 302, a
convolution operation is applied. The convolution layer 304
consists of multiple kernels k with varied sizes. Each kernel k
with size s.times.w is applied to a subinterval that contains s
weeks, represented as:
w.sub.i:w.sub.i+s (Formula 3)
From this, we may represent a new feature f.sub.i using the
following expression:
f.sub.i=C(kw.sub.i:w.sub.i+s+b) (Formula 4)
Based on this expression, first, the dot product of kernel k and
weeks w.sub.i to w.sub.i+s is calculated. Then, the result is added
up with b, the bias term. Finally, a non-linear function C, such as
RELU, is applied. The whole operation is applied to an interval
(h-s+1) times, and generates the feature map F:
F={f.sub.1, f.sub.2, . . . f.sub.h-s+1} (Formula 5)
[0085] The next layer of the CNN model is the pooling layer 306.
The main task of the pooling layer 306 is capturing prominent
feature(s) from a feature map. The pooling operation can be applied
in two different approaches: global or local. A global pooling
operation acts as an aggregate function and converts a feature map
to a value:
v=Arg.sub.op({f.sub.1, f.sub.2, . . . f.sub.h-s+1}) (Formula 6)
Depending on the domain of an application, either maximum functions
or averaging functions may be used as an operator (op). A local
pooling operation slides through a feature map and aggregates m
values of each window to produce a set of values v as set forth by
the following formula:
v={Arg.sub.op(f.sub.i:f.sub.i+m}) .sym. . . . .sym.
Arg.sub.op({f.sub.h-s+1-m:f.sub.h-s+1})} (Formula 7)
[0086] Applying all filters and concatenating new extracted
features results in a one-dimensional array, called V. This is the
input to the next layer, which is a full connected RELU layer 308,
to which a softmax squashing function may be applied, shown
generally below:
.sigma. ( z ) j = e z j k = 1 K e z k ( Formula 8 )
##EQU00001##
where z is a vector of the inputs to the output layer 310, with j
indexes of the output units. The output of softmax function, which
include values between 0 and 1, passes to the final output layer
310 and gives a probability p. When building the neural network
predictive model based on historic data, p can be compared to a
known (or predetermined label) value of 0 or 1 to calculate an
error rate. Based on this comparison, the model may update network
parameters including weights and biases of the convolution layers.
When creating the initial predictive model, the parameters and
biases may be initiated randomly and adjusted/learned through
successive iterations. This provides the benefit of a
self-improving predictive model.
[0087] With specific reference to FIG. 7, and in some non-limiting
embodiments or aspects, the CNN predictive model architecture 300
may be reconfigured to pass the one-dimensional extracted feature
array V to an LSTM layer 312 instead of the fully-connected RELU
layer 308. LSTM may be used to aggregate the temporal aspects in
the features learned from the convolutional network. LSTM takes
input in the form of a sequence to compute a hidden vector sequence
and an output vector by iterating the following equations from t=1
to H:
h.sub.t=H{W.sub.vhv.sub.t+W.sub.hhh.sub.t-1+b.sub.h} (Formula
9)
o.sub.t=H{W.sub.hoh.sub.t+b.sub.y} (Formula 10)
where the W terms denote weight matrices, b terms denote bias
vectors, and H is the hidden layer function. FIG. 7 depicts this
CNN-LSTM predictive model architecture.
[0088] Although the disclosure has been described in detail for the
purpose of illustration based on what is currently considered to be
the most practical and preferred and non-limiting embodiments, it
is to be understood that such detail is solely for that purpose and
that the disclosure is not limited to the disclosed embodiments,
but, on the contrary, is intended to cover modifications and
equivalent arrangements that are within the spirit and scope of the
appended claims. For example, it is to be understood that the
present disclosure contemplates that, to the extent possible, one
or more features of any embodiment can be combined with one or more
features of any other embodiment.
* * * * *